The science

Working memory in early childhood, and what every parent should know

By Ryan King · Updated June 2026 · Sourced.

Working memory is the cognitive workspace where a child holds a few things in mind while doing something with them. It grows quickly between ages three and seven. The way you talk and play affects how much it grows.

What working memory is

Working memory is the small mental workspace where a person holds and manipulates information for a few seconds. It is different from short-term memory, which is passive holding. Working memory does work: holding three numbers in mind while adding them, listening to a sentence while parsing its meaning, planning the next move in a game while remembering the rules.

Alan Baddeley's model is the standard framework. It distinguishes a phonological loop for verbal information, a visuospatial sketchpad for images and locations, and a central executive that coordinates them. Each component grows on its own developmental timeline.

Why parents should care

Working memory at ages 4 to 6 is one of the strongest single predictors of academic outcomes by age 11, including reading and math. Tracy Alloway's longitudinal research finds it more predictive than IQ in some studies. The mechanism is intuitive: a child whose working memory is too small to hold a multi-step instruction cannot follow it, regardless of whether they understood each step.

It is also the executive function that supports self-regulation. A child who can hold 'I want the cookie, but I am supposed to wait until after dinner' in mind for sixty seconds is doing working memory work.

What grows it

Active play that demands holding several things in mind: card matching games, scavenger hunts with multiple targets, sequence-recall games, the cup-shuffle game where the ball moves under cups while the child watches. Reading aloud and asking the child to predict what comes next. Cooking together, where the child has to remember which ingredient comes after the next step.

The boring fact is that any structured play that requires the child to hold a thread for a while exercises working memory. Apps that train it deliberately exist, and the research on whether the gains transfer outside the trained task is genuinely mixed. The safer bet is the broad set of activities, not the narrow program.

What undermines it

Rapid-paced media is the leading suspect, particularly in younger children. Constant context switching is the same. A child who is interrupted during play five times before they finish what they were doing does not get the practice that supports holding a longer thread next time.

Sleep is the unsexy answer that always comes up. Working memory performance drops measurably in under-rested children. So does the recently popular concept of 'pretend play', which is itself working memory exercise.

What we are building

Cairn Memory is the second app on the roadmap. It targets working memory directly through the activity types the research supports: phonological memory with letter and word sequences, visuospatial memory with tile-based memory games, and a Simon-like sequencing activity that scales with the child's age. The pedagogy note inside the app cites the Baddeley framework and the developmental norms.

We are deliberately not promising 'IQ gains'. The honest claim is that practicing the activities the research describes for working memory should help a child do the activities the research describes for working memory. Whether and how far that transfers to school is something to watch, not to over-promise.

Sources: Baddeley (2000), The Episodic Buffer, Trends in Cognitive Sciences; Alloway and Alloway (2010) on working memory and academic attainment; Diamond and Lee (2011), Science, on executive function training.